Atmospheric distillation column control method, device, electronic equipment and storage medium

Abstract

The application discloses a kind of normal pressure distillation column control method, device, electronic equipment and storage medium.Therein, the method comprises: for each parameter control moment in the operation process of the distillation column to be controlled, obtain the distillation process data corresponding to the distillation column to be controlled at the current parameter control moment;Based on the operation parameter prediction model determined in advance, the distillation process data corresponding to the current parameter control moment is processed, and the parameter adjustment amount of at least one operation parameter corresponding to the current parameter control moment of the distillation column to be controlled is obtained;Based on the parameter adjustment amount of at least one operation parameter and the parameter adjustment mode corresponding to the current parameter control moment, the operation parameter is adjusted.This technical solution realizes the effect of accurately controlling the operation parameter of the distillation column during the operation process of the distillation column, achieves the effect of reducing the labor cost and ensuring the stability of the distillation column system, and also improves the operation efficiency of the distillation column.

Description

Technical Field

[0001] This invention relates to the field of petroleum refining technology, and in particular to a method, apparatus, electronic device and storage medium for controlling an atmospheric distillation tower. Background Technology

[0002] Atmospheric distillation is the first step in crude oil processing and one of the most important units in oil refining plants. In industrial production, fluctuations often occur due to various factors such as raw material composition and feed load, affecting production safety and product quality.

[0003] In related technologies, feedforward technology is commonly used to control atmospheric distillation processes. Feedforward technology uses a PID controller to control the reboiler.

[0004] However, feedforward techniques typically require the establishment of complex mathematical models, which not only incurs high costs but also limits their flexibility and portability across different scenarios. If the system's dynamic model changes, previously tuned PID parameters become unusable. Furthermore, most PID parameter adjustments at present are based on manual tuning or expert systems, both of which rely heavily on manual parameter selection and domain knowledge, significantly reducing the ease of use, flexibility, and versatility of PID controllers. Summary of the Invention

[0005] This invention provides a method, device, electronic equipment, and storage medium for controlling an atmospheric distillation column, so as to achieve precise control of the operating parameters of the distillation column during operation, thereby reducing labor costs, ensuring the stability of the distillation column system, and improving the operating efficiency of the distillation column.

[0006] According to one aspect of the present invention, a method for controlling an atmospheric distillation column is provided, the method comprising:

[0007] For each parameter control time during the operation of the distillation column to be controlled, the distillation process data corresponding to the current parameter control time of the distillation column to be controlled is obtained. The distillation process data includes at least one distillation process sequence. The at least one distillation process sequence includes the column temperature sequence and column pressure sequence, column liquid level sequence corresponding to each preset column height, the quality parameter sequence corresponding to at least one purification object, and the liquid flow rate sequence corresponding to at least one target pipeline. Each sequence value included in each distillation process sequence corresponds to each parameter control time.

[0008] Based on a pre-determined operating parameter prediction model, the distillation process data corresponding to the current parameter control time is processed to obtain the parameter adjustment amount of at least one operating parameter of the distillation column under control at the current parameter control time. The operating parameter prediction model is obtained by training based on a reinforcement learning algorithm. The at least one operating parameter includes valve opening, reflux flow rate, and / or the extraction amount of each of the purified objects.

[0009] The operating parameters are adjusted based on the parameter adjustment amount of the at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time.

[0010] According to another aspect of the present invention, an atmospheric distillation column control device is provided, the device comprising:

[0011] The data acquisition module is used to acquire distillation process data corresponding to the current parameter control time of the distillation column under control during the operation of the distillation column under control. The distillation process data includes at least one distillation process sequence, which includes a column temperature sequence, a column pressure sequence, a column liquid level sequence, a quality parameter sequence corresponding to at least one purification object, and a liquid flow rate sequence corresponding to at least one target pipeline. Each sequence value included in each distillation process sequence corresponds to the control time of each parameter.

[0012] The data processing module is used to process the distillation process data corresponding to the current parameter control time based on a pre-determined operation parameter prediction model, to obtain the parameter adjustment amount of at least one operation parameter of the distillation column to be controlled at the current parameter control time, wherein the operation parameter prediction model is trained based on a reinforcement learning algorithm; the at least one operation parameter includes valve opening, reflux flow rate and / or the extraction amount of each of the purification objects;

[0013] The parameter adjustment module is used to adjust the operating parameters based on the parameter adjustment amount of the at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time.

[0014] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0015] At least one processor; and

[0016] A memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the atmospheric distillation column control method according to any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the atmospheric distillation column control method according to any embodiment of the present invention.

[0019] The technical solution of this invention obtains distillation process data corresponding to the current parameter control time of the distillation column during its operation by controlling the control time of each parameter. Further, based on a pre-determined operating parameter prediction model, the distillation process data corresponding to the current parameter control time is processed to obtain the parameter adjustment amount of at least one operating parameter of the distillation column at the current parameter control time. Finally, the operating parameter is adjusted based on the parameter adjustment amount of at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time. This solves the problems of inaccurate control of the distillation process and complex control processes in related technologies, achieving precise control of the distillation column's operating parameters during operation. This results in reduced labor costs and improved distillation column operating efficiency while ensuring system stability.

[0020] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of an atmospheric distillation column control method according to Embodiment 1 of the present invention;

[0023] Figure 2 This is a flowchart of an atmospheric distillation column control method according to Embodiment 2 of the present invention;

[0024] Figure 3 This is a schematic diagram of the structure of an atmospheric distillation column control device according to Embodiment 3 of the present invention;

[0025] Figure 4 This is a schematic diagram of the structure of an electronic device for implementing the atmospheric distillation column control method of this invention. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0028] Example 1

[0029] Figure 1 This is a flowchart of an atmospheric distillation column control method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where the operating parameters of a distillation column are controlled during operation. This method can be executed by an atmospheric distillation column control device, which can be implemented in hardware and / or software and can be configured in a terminal and / or server. Figure 1 As shown, the method includes:

[0030] S110. For the control time of each parameter in the distillation process of the distillation column to be controlled, obtain the distillation process data corresponding to the current parameter control time of the distillation column to be controlled.

[0031] It should be noted that the technical solutions of this invention can be applied to the distillation process control of an atmospheric distillation column. During the stable operation of an atmospheric distillation column, fluctuations often occur due to various factors such as raw material composition and feed load, affecting production safety and product quality. To address these fluctuations, at least one operating parameter associated with the distillation column can be controlled during operation, such as valve opening, reflux flow rate, and product extraction rate. This ensures stable operation of the atmospheric distillation column and that the separated product meets quality requirements.

[0032] In this embodiment, the distillation column to be controlled can be understood as an atmospheric distillation column to be controlled. Atmospheric distillation columns are used in equipment in fields such as petrochemicals. Their distillation principle involves partially vaporizing a liquid mixture, utilizing the different volatility of its components to achieve separation. The bottom of the atmospheric distillation column is liquid, and gas is distilled from the top. The main function of an atmospheric distillation column is to separate mixed liquids. Under atmospheric pressure, different components have different boiling points. Therefore, the components can be gradually boiled by heating, and then condensed into liquid by cooling, thus achieving purification. Parameter control time can be understood as the time during which the relevant control parameters of the distillation column are adjusted during its operation. In practical applications, the distillation column operation process can be divided into multiple time periods according to a preset step size, and these divided time periods are used as parameter control time periods. The preset step size can be any value, selectable, such as 5 minutes or 10 minutes, etc.

[0033] In this embodiment, distillation process data can be understood as the data generated by the distillation column under control when distilling the mixed liquid inside the column. Distillation process data characterizes the distillation process; therefore, it can be a sequence of parameter values ​​characterizing the operating state of the distillation column corresponding to different parameter control times. The distillation process data may include at least one distillation process sequence, which includes a column temperature sequence and a column pressure sequence corresponding to each preset column height, a column liquid level sequence, a quality parameter sequence corresponding to at least one purification target, and a liquid flow rate sequence corresponding to at least one target pipeline. Each sequence value included in each distillation process sequence corresponds to a parameter control time.

[0034] In this context, each preset column height can be understood as a different height within the distillation column to be controlled. The column temperature sequence can be understood as the temperature value corresponding to different parameter control times at each height in the distillation column to be controlled. The column pressure sequence can be understood as the pressure value corresponding to different parameter control times at each height in the distillation column to be controlled. In practical applications, multiple different heights can be predetermined within the distillation column to be controlled, and temperature and pressure detection devices can be installed at each height. During the distillation process, the temperature value at each height under different parameter control times is detected by the temperature detection device to obtain the column temperature sequence corresponding to that height; simultaneously, the pressure value at each height under different parameter control times is detected by the pressure detection device to obtain the column pressure sequence corresponding to that height. The column liquid level height sequence can include the liquid level height within the column corresponding to each parameter control time. The liquid within the column can be the mixed liquid within the distillation column to be controlled, i.e., the liquid to be extracted. The purification object can be understood as the object from which the mixed liquid is purified. For example, assuming the mixed liquid to be purified is crude oil, the purification object can be gasoline, diesel, or lubricating oil, etc. A quality parameter sequence can be understood as a sequence of parameters characterizing the quality features of the corresponding purified object. It should be noted that the corresponding quality characteristic parameters differ for different extracted objects. For example, if the purified object is kerosene, the quality characteristic parameters are flash point and / or 90% distillation point, and the corresponding quality parameter sequence can be a flash point sequence and / or a 90% distillation point sequence; if the purified object is diesel oil, the quality characteristic parameter is 90% (or 95%) point, and the corresponding quality parameter sequence can be a 90% point sequence or a 95% point sequence. Those skilled in the art will understand that flash point is the lowest temperature at which a material product, when mixed with ambient air, ignites and immediately burns upon contact with a flame. Flash point is a safety indicator for the storage, transportation, and use of flammable liquids, and also an indicator of the volatility of flammable liquids. The target pipeline can be a pipeline associated with the distillation process in the distillation column to be controlled. The target pipeline can be at least a portion of all pipelines installed in the distillation column to be controlled. The liquid flow rate sequence can include the liquid flow rate corresponding to the target pipeline at each parameter control time.

[0035] In practical applications, to ensure stable operation of the controlled distillation column and high-quality purified product, parameters characterizing the distillation state can be monitored during the distillation of the mixed liquid. Based on the monitoring results, a corresponding parameter control strategy for the controlled distillation column can be determined. Therefore, during the distillation of the mixed liquid in the controlled distillation column, distillation data can be collected at each parameter control moment. This allows for the acquisition of distillation process data corresponding to the current parameter control moment.

[0036] S120. Based on a pre-determined operating parameter prediction model, process the distillation process data corresponding to the current parameter control time to obtain the parameter adjustment amount of at least one operating parameter of the distillation column to be controlled at the current parameter control time.

[0037] The operational parameter prediction model is trained based on a reinforcement learning algorithm.

[0038] In this embodiment, the operating parameter prediction model can be understood as a policy model determined by a reinforcement learning algorithm. This policy model can be used to determine the parameter adjustment value of at least one operating parameter of the distillation column under control. The policy model can be a deep neural network model that deploys a policy function. The input object of this model can be the state of the agent's environment, and the output object can be the decision action determined based on the input state. The operating parameter prediction model can be a neural network model with any model structure. Optionally, the model structure of the operating parameter prediction model can include, but is not limited to, the model structure of a Gated Recurrent Neural Network (GRU) and a Long Short-Term Memory Network (LSTM). The operating parameter can be the parameter to be adjusted and controlled during the distillation process of the distillation column under control. At least one operating parameter can include valve opening, reflux flow rate, and the extraction amount of each purified object. Valve opening refers to the relative position of the valve between the open and closed states. The size of the valve opening directly affects the flow rate and resistance of the fluid, playing an important role in fluid control and regulation. Reflux flow rate can be understood as the flow rate at which a portion of the extracted material is reintroduced into the distillation column during the purification process of a mixed liquid. Extraction rate can be understood as the amount of liquid extracted from the total amount of the purified material. For each operating parameter, the adjustment amount of the current operating parameter can be understood as the increase or decrease in the parameter value of the current operating parameter; that is, the increase or decrease in the parameter value based on the value of the operating parameter at the current control moment. The adjustment amount of the current operating parameter can be a quantity with both direction and magnitude, with direction represented by "+" and "-".

[0039] In practical applications, to determine the parameter adjustment amounts for each operating parameter of the distillation column under control at the current parameter control time, the distillation process data at the current parameter control time can be obtained and input into the operating parameter prediction model. Furthermore, the distillation process data can be processed based on the operating parameter prediction model to obtain the parameter adjustment amounts for valve opening, reflux flow rate, and / or the extraction amounts of each purification target at the current parameter control time.

[0040] S130. Adjust the operating parameter based on the parameter adjustment amount of at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time.

[0041] In this embodiment, for each parameter control time during the operation of the distillation column to be controlled, after obtaining the parameter adjustment amount of at least one operating parameter corresponding to the current parameter control time, the corresponding operating parameter at the current parameter control time can be adjusted according to the predetermined parameter adjustment method and the parameter adjustment amount of at least one operating parameter corresponding to the current parameter control time.

[0042] In this embodiment, the parameter adjustment method can be understood as the method of adjusting the operating parameters of the distillation column during operation. The parameter adjustment method can be any method capable of adjusting the operating parameters. Optionally, the parameter adjustment method can include direct adjustment or indirect adjustment, etc.

[0043] In practical applications, after obtaining the parameter adjustment amount of at least one operating parameter of the distillation column under control at the current parameter control time, in order to ensure the stable operation of the distillation column under control while achieving automatic parameter control, it is also possible to determine whether the parameter value of the operating parameter after parameter adjustment reaches the preset control threshold. Then, the corresponding parameter adjustment method can be determined based on the judgment result.

[0044] Based on this, in addition to the above technical solutions, the method further includes: determining the parameter values ​​of each operating parameter at the next parameter control time based on the parameter adjustment amount of at least one operating parameter at the current parameter control time and the parameter values ​​of each operating parameter at the current parameter control time obtained in advance; determining the parameter adjustment method at the current parameter control time as direct adjustment when none of the parameter values ​​of each operating parameter at the next parameter control time reach the corresponding preset control threshold; and determining the parameter adjustment method at the current parameter control time as indirect adjustment when one of the operating parameters reaches the corresponding preset control threshold in the next parameter control time.

[0045] In this embodiment, the preset control threshold can be any value, and different preset control thresholds can correspond to different operating parameters.

[0046] In practical applications, after obtaining the parameter adjustment amount corresponding to at least one operating parameter at the current parameter control time, for each operating parameter, the parameter adjustment amount corresponding to the current operating parameter at the current parameter control time can be added to the pre-acquired parameter value corresponding to the current operating parameter at the current parameter control time. This allows the acquisition of the parameter value corresponding to the current operating parameter at the next parameter control time. Furthermore, after obtaining the parameter values ​​corresponding to each operating parameter at the next parameter control time, each parameter value can be compared with the corresponding preset control threshold. If it is determined that none of the parameter values ​​have reached the corresponding preset control threshold, the parameter adjustment method corresponding to the current parameter control time can be determined as direct adjustment. If it is determined that one of the parameter values ​​has reached the corresponding preset control threshold, the parameter adjustment method corresponding to the current parameter control time can be determined as indirect adjustment.

[0047] Furthermore, once the parameter adjustment method corresponding to the current parameter control moment is determined, the operating parameters corresponding to the current parameter control moment can be adjusted according to the parameter adjustment method and the parameter adjustment amount of each operating parameter.

[0048] Optionally, the parameter adjustment method is direct adjustment, which adjusts the operating parameters at the corresponding parameter control time according to a predetermined parameter adjustment method and the parameter adjustment amount of at least one operating parameter corresponding to each parameter control time. This includes: adjusting the valve opening of the distillation column under control based on the parameter adjustment amount corresponding to the current parameter control based on the valve opening; and / or adjusting the reflux flow rate of the distillation column under control based on the parameter adjustment amount corresponding to the current parameter control based on the reflux flow rate; and / or adjusting the extraction amount of the corresponding purification object based on the parameter adjustment amount of the extraction amount of each purification object.

[0049] In this embodiment, direct adjustment can be understood as directly adjusting the operating parameters, that is, adjusting the operating parameters of the distillation column under control without human intervention.

[0050] In practical applications, after obtaining the parameter adjustment values ​​for each operating parameter at the current parameter control moment, the operating parameters at the current parameter control moment can be adjusted based on these adjustment values. Specifically, the valve opening of the distillation column under control can be adjusted based on the direction and magnitude of the adjustment value corresponding to the current parameter control. Furthermore, the reflux flow rate of the distillation column under control can be adjusted based on the direction and magnitude of the adjustment value corresponding to the current parameter control. Additionally, the extraction amount of each purified object can be adjusted based on the parameter adjustment value for the extraction amount of each purified object.

[0051] Optionally, the parameter adjustment method is indirect adjustment. The operation parameters at the corresponding parameter control time are adjusted according to a predetermined parameter adjustment method and the parameter adjustment amount of at least one operation parameter corresponding to each parameter control time. This includes: displaying the parameter adjustment amount of each operation parameter at the current parameter control time on the display interface of the target terminal; in response to an edit trigger operation for at least one parameter editing item, determining the target adjustment amount corresponding to each operation parameter, and adjusting the corresponding operation parameter based on each target adjustment amount.

[0052] In this embodiment, indirect adjustment can be understood as adjusting the operating parameters based on parameter editing trigger operations, that is, adjusting the corresponding operating parameters based on the parameter adjustment amount input on the display interface. The target terminal can be understood as a terminal device used to process distillation process data, or as a device used to deploy an operating parameter prediction model. Optionally, the target terminal can be a fixed terminal or a mobile terminal; this embodiment does not specifically limit this. The display interface can show parameter editing items corresponding to each operating parameter. Parameter editing items can be controls for editing operating parameters. The target adjustment amount is the value input in the parameter editing item.

[0053] In practical applications, after obtaining the parameter adjustment values ​​corresponding to each operation parameter at the current parameter control moment, these adjustment values ​​can be displayed on the target terminal's display interface, allowing users to use them as a basis for editing. Furthermore, upon detecting an edit trigger operation on a parameter editing item, the value entered in the parameter editing item can be obtained and used as the target adjustment value for the corresponding operation parameter. Then, the corresponding operation parameter can be adjusted based on these target adjustment values.

[0054] The technical solution of this invention obtains distillation process data corresponding to the current parameter control time of the distillation column during its operation by controlling the control time of each parameter. Further, based on a pre-determined operating parameter prediction model, the distillation process data corresponding to the current parameter control time is processed to obtain the parameter adjustment amount of at least one operating parameter of the distillation column at the current parameter control time. Finally, the operating parameter is adjusted based on the parameter adjustment amount of at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time. This solves the problems of inaccurate control of the distillation process and complex control processes in related technologies, achieving precise control of the distillation column's operating parameters during operation. This results in reduced labor costs and improved distillation column operating efficiency while ensuring system stability.

[0055] Example 2

[0056] Figure 2 This is a flowchart of an atmospheric distillation column control method provided in Embodiment 2 of the present invention. Based on the foregoing embodiments, an operating parameter prediction model can be trained before processing the distillation process data based on the operating parameter prediction model. Specific implementation methods can be found in the technical solution of this embodiment. Technical terms that are the same as or similar to those in the above embodiments will not be repeated here.

[0057] like Figure 2 As shown, the method includes:

[0058] S210, Train to obtain the operational parameter prediction model.

[0059] It should be noted that before applying the operation parameter prediction model provided in this embodiment of the invention, a pre-built neural network model needs to be trained first to obtain the operation parameter prediction model. Before training the model, multiple online training samples can be constructed, and then the model can be trained based on the online training samples.

[0060] Optionally, training the operational parameter prediction model includes: for each training round, obtaining the online training samples corresponding to the distillation column to be controlled in the current training round; and training the operational parameter prediction model based on each online training sample and the reinforcement learning algorithm.

[0061] In this embodiment, a training round can be understood as a training process, that is, the process from the initial state of the environment model until the preset maximum number of training steps is reached. After each training round, the environment model can be initialized to restore it to its initial state. Then, the next training round can be executed. Furthermore, after training for multiple rounds and when the loss function in the model is detected to have converged, the trained model can be obtained. In this embodiment, the initial state can be any state in the distillation process of the controlled distillation column. That is, in practical applications, an operating parameter prediction model can be introduced at any time in the distillation process of the controlled distillation column to adjust the distillation process of the controlled distillation column based on the operating parameter prediction model.

[0062] The online training samples are determined based on the environmental model corresponding to the distillation column to be controlled.

[0063] In this embodiment, the environment model can be understood as a simulation model representing the distillation process of the controlled distillation column, or as an environment model obtained by modeling the actual operating environment of the controlled distillation column. It can be understood that the finally trained operating parameter prediction model is a strategy model applied to the actual distillation process of the controlled distillation column, and this strategy model is trained based on a reinforcement learning algorithm. During the training process of the operating parameter prediction model, in order to reduce the number of trial-and-error samples in the real environment, an environment model corresponding to the real environment can be constructed. Therefore, a large number of trial-and-error samples during the training process can be completed within the environment model.

[0064] In this embodiment, online training samples can be understood as online sample data collected in real time during the operation of the environmental model. The online training samples include at least one distillation process sample sequence. Each distillation process sample sequence includes a sample sequence of temperature and pressure inside the column at each preset column height, a sample sequence of liquid level height inside the column, a sample sequence of quality parameters corresponding to at least one purification target, a sample sequence of liquid flow rate in at least one target pipeline, and a reward feedback information sequence. The values ​​of each sample sequence included in each distillation process sample sequence correspond to the control time of each parameter.

[0065] The preset column height can be any height within the distillation column to be controlled. The column temperature sample sequence can include column temperature sample values ​​corresponding to each parameter control time. These column temperature sample values ​​can be the temperature values ​​at the corresponding column height determined based on the environmental model. The column pressure sample sequence can include column pressure sample values ​​corresponding to each parameter control time. These column pressure sample values ​​can be the pressure values ​​at the corresponding column height determined based on the environmental model. The column liquid level height sample sequence can include column liquid level height sample values ​​corresponding to each parameter control time. These column liquid level height sample values ​​can be the liquid level height values ​​within the distillation column determined based on the environmental model. The quality parameter sample sequence can include quality parameter sample values ​​of the corresponding purified object corresponding to each parameter control time. These quality parameter sample values ​​can be the quality parameter values ​​of the corresponding purified object determined based on the environmental model. The liquid flow rate sample sequence can include liquid flow rate sample values ​​corresponding to each parameter control time. These liquid flow rate sample values ​​are the flow rates of the liquid in the corresponding target pipeline determined based on the environmental model. The reward feedback information sequence can include reward feedback information corresponding to each parameter control time. Reward feedback is a numerical value obtained by an agent after performing an action, which can characterize the quality of the action. Reward feedback can be determined based on a reward function deployed in the environment model.

[0066] In practical applications, an environmental model corresponding to the distillation process of the distillation column to be controlled can be pre-constructed. Then, for each training round, scenario initialization can be performed to bring the environmental model to its initial state. Next, the environmental model is run, and during its operation, distillation process data is collected at each parameter control moment. This allows for the acquisition of sample values ​​of the column temperature and pressure at each preset column height at each parameter control moment, and the construction of column temperature and pressure sample sequences at each preset column height based on the collected temperature sample values. Simultaneously, sample values ​​of the column liquid level height at each parameter control moment can be obtained, and a column liquid level height sample sequence can be constructed based on these sample values. Furthermore, sample values ​​of the quality parameters of each purified object at each parameter control moment can be obtained, and a quality parameter sample sequence can be constructed based on these sample values. Finally, sample values ​​of the liquid flow rate of each target pipeline at each moment can also be obtained, and a sample sequence of liquid flow rate values ​​for each target pipeline can be obtained.

[0067] Furthermore, for each parameter control moment, after determining the parameter adjustment amount of at least one operating parameter corresponding to the current parameter control moment based on the simulation model, and adjusting the corresponding operating parameter based on the parameter adjustment amount, the state information of the environment model after parameter adjustment is processed based on the reward function pre-deployed in the environment model to obtain the reward feedback information corresponding to the current parameter control moment. Then, a reward feedback information sequence can be constructed based on the reward feedback information corresponding to each parameter control moment.

[0068] Furthermore, the operational parameter prediction model can be trained based on various online training samples and reinforcement learning algorithms to obtain a trained operational parameter prediction model. The training process of the operational parameter prediction model will be explained in detail below.

[0069] First, for each online training sample, the sample sequences of tower temperature, tower pressure, tower liquid level, quality parameters of at least one purification object, and liquid flow rate of at least one target pipeline corresponding to each preset tower height in the current online training sample are input into the operating parameter prediction model to obtain the probability distribution of the adjustment amount of the distillation tower under control at the current parameter control time.

[0070] The regulation probability distribution includes at least one actual regulation probability set.

[0071] In this embodiment, the actual adjustment probability set includes the actual adjustment amount of each operating parameter and the corresponding probability information. The actual adjustment amount of each operating parameter may include the actual adjustment amount of valve opening, the actual adjustment amount of reflux flow rate, and / or the actual adjustment amount of extraction of each purified object. The probability information can be used to characterize the likelihood of the actual adjustment amount of the corresponding operating parameter.

[0072] In practical applications, for each online training sample, the sample sequences of tower temperature, tower pressure, tower liquid level, quality parameters corresponding to at least one purification object, and liquid flow rate corresponding to at least one target pipeline can be input into the operating parameter prediction model. The input data can then be processed based on the operating parameter prediction model, and the probability distribution of the adjustment amount corresponding to the distillation tower under control at the current parameter control time can be output.

[0073] Subsequently, for each actual adjustment probability set in the adjustment probability distribution, the actual adjustment of each operating parameter in the current actual adjustment probability set, the sample sequences of temperature and pressure in the tower corresponding to each preset tower height, the sample sequence of liquid level in the tower, the sample sequence of quality parameters corresponding to at least one purification object, and the sample sequence of liquid flow in at least one target pipeline are input into the state-action value model to obtain the expected reward corresponding to the current actual adjustment probability set.

[0074] In this embodiment, the state-action value network can be a neural network that includes a state-action value function. The input to the state-action value network can be the current state and the current decision action, and the output can be the value corresponding to taking the decision action for the current state, that is, the expected reward corresponding to the decision action at the current moment.

[0075] In practical applications, after obtaining the probability distribution of the adjustment amount, in order to evaluate each actual adjustment amount probability set included in the probability distribution and determine the value of taking the decision action corresponding to each actual adjustment amount probability set under the current parameter control time and the corresponding state of the controlled distillation column environment model, the actual adjustment amount of each operating parameter in the current actual adjustment amount probability set, the sample sequences of column temperature and pressure at each preset column height, the sample sequence of column liquid level, the sample sequence of quality parameters corresponding to at least one purification object, and the sample sequence of liquid flow rate in at least one target pipeline are input into the state-action value model to obtain the expected reward corresponding to the current actual adjustment amount probability set.

[0076] Subsequently, the parameters of the operational parameter prediction model are updated based on the adjustment probability distribution, expected rewards, and policy gradient algorithms; and the target actual adjustment probability set is determined based on the adjustment probability distribution.

[0077] Policy gradient algorithms are a class of algorithms for solving reinforcement learning problems. They are gradient-based optimization algorithms that help machine learning models optimize in decision-making environments to achieve the best results. The idea behind policy gradient algorithms is to first represent the policy as a continuous function related to reward, and then use optimization methods for continuous functions to find the optimal policy. The optimization objective is to maximize the continuous function.

[0078] In practical applications, after obtaining the expected reward corresponding to each actual regulation probability set in the regulation probability distribution, the regulation probability distribution and each expected reward can be processed using the policy gradient algorithm to update the model parameters of the operating parameter prediction model. Furthermore, the regulation probability distribution can be sampled, and the target actual regulation probability set can be determined from each actual regulation probability set in the regulation probability distribution to obtain the actual regulation corresponding to each operating parameter.

[0079] Subsequently, based on the pre-set reward function, the sample sequences of temperature, pressure, liquid level, quality parameters of at least one purification target, liquid flow rate in at least one target pipeline, and actual adjustment of each operating parameter in the target actual adjustment probability set are processed to determine the reward feedback information corresponding to the distillation column under control at the current parameter control time.

[0080] The reward function can be understood as a function used to evaluate the state and decision-making action at any given time. The reward function can vary depending on the application scenario; that is, it can be set according to actual needs. In this embodiment, the reward function may include a safety reward function for observation indicators, a stability reward function for observation indicators, an anti-interference reward function for control indicators, and / or a regional value attribute reward function. Optionally, the reward function can be determined by a weighted sum of the safety reward function for observation indicators, the stability reward function for observation indicators, the anti-interference reward function for control indicators, and the regional value attribute reward function. Observation indicators include sample values ​​of temperature and pressure inside the tower at each preset tower height, sample values ​​of liquid level height inside the tower, sample values ​​of quality parameters corresponding to each purified object, and sample values ​​of liquid flow rate in each target pipeline. Control indicators may include valve opening, reflux flow rate, and / or the extraction amount of each purified object.

[0081] In practical applications, after obtaining the probability set of the target actual adjustment amount, the actual adjustment amount of each operating parameter in the probability set of the target actual adjustment amount, the sample sequence of the temperature inside the tower and the sample sequence of the pressure inside the tower corresponding to each preset tower height, the sample sequence of the liquid level height inside the tower, the sample sequence of the quality parameter corresponding to at least one purification object, and the sample sequence of the liquid flow rate corresponding to at least one target pipeline can be input into the pre-set reward function. Then, the reward feedback information corresponding to the distillation tower under control at the current parameter control time can be obtained.

[0082] Subsequently, based on the reward feedback information and the temporal difference algorithm, the parameters of the state-action value model are updated, and the training ends when the preset training target corresponding to the operation parameter prediction model is reached, based on the reward feedback information.

[0083] Temporal Difference (TD) is an algorithm used to estimate the value function of a policy. It adaptively adjusts the policy by learning the difference between the current state and future states. The core idea of ​​the TD algorithm is the updating of the state value function. The preset training objective can be a pre-set termination condition for the policy network training process. Optionally, the preset training objective can include maximizing the objective function value corresponding to the policy gradient algorithm or reaching a preset number of training iterations.

[0084] In practical applications, after obtaining reward feedback information, the parameters of the state-action value model can be updated based on the reward feedback information and the temporal difference algorithm, and the reward feedback information sequence can also be updated based on the reward feedback information. Then, training ends when the preset training objective corresponding to the operation parameter prediction model is reached, resulting in the operation parameter prediction model.

[0085] S220. For the control time of each parameter in the distillation process of the distillation column to be controlled, obtain the distillation process data corresponding to the current parameter control time of the distillation column to be controlled.

[0086] S230. Based on a pre-determined operating parameter prediction model, process the distillation process data corresponding to the current parameter control time to obtain the parameter adjustment amount of at least one operating parameter of the distillation column to be controlled at the current parameter control time.

[0087] S240. Adjust the operating parameter based on the parameter adjustment amount of at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time.

[0088] The technical solution of this invention obtains distillation process data corresponding to the current parameter control time of the distillation column during its operation by controlling the control time of each parameter. Further, based on a pre-determined operating parameter prediction model, the distillation process data corresponding to the current parameter control time is processed to obtain the parameter adjustment amount of at least one operating parameter of the distillation column at the current parameter control time. Finally, the operating parameter is adjusted based on the parameter adjustment amount of at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time. This solves the problems of inaccurate control of the distillation process and complex control processes in related technologies, achieving precise control of the distillation column's operating parameters during operation. This results in reduced labor costs and improved distillation column operating efficiency while ensuring system stability.

[0089] Example 3

[0090] Figure 3 This is a schematic diagram of the structure of an atmospheric distillation column control device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: a data acquisition module 310, a data processing module 320, and a parameter adjustment module 330.

[0091] The data acquisition module 310 is used to acquire distillation process data corresponding to the current parameter control time of the distillation column during its operation, wherein the distillation process data includes at least one distillation process sequence, which includes a column temperature sequence, a column pressure sequence, a column liquid level sequence, a quality parameter sequence corresponding to at least one purification object, and a liquid flow rate sequence corresponding to at least one target pipeline, and each sequence value included in each distillation process sequence corresponds to a parameter control time; the data processing module 320 is used to process the distillation process data corresponding to the current parameter control time based on a pre-determined operating parameter prediction model to obtain the parameter adjustment amount of at least one operating parameter of the distillation column corresponding to the current parameter control time, wherein the operating parameter prediction model is trained based on a reinforcement learning algorithm; the at least one operating parameter includes valve opening, reflux flow rate, and / or the extraction amount of each purification object; the parameter adjustment module 330 is used to adjust the operating parameter based on the parameter adjustment amount of the at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time.

[0092] The technical solution of this invention obtains distillation process data corresponding to the current parameter control time of the distillation column during its operation by controlling the control time of each parameter. Further, based on a pre-determined operating parameter prediction model, the distillation process data corresponding to the current parameter control time is processed to obtain the parameter adjustment amount of at least one operating parameter of the distillation column at the current parameter control time. Finally, the operating parameter is adjusted based on the parameter adjustment amount of at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time. This solves the problems of inaccurate control of the distillation process and complex control processes in related technologies, achieving precise control of the distillation column's operating parameters during operation. This results in reduced labor costs and improved distillation column operating efficiency while ensuring system stability.

[0093] Optionally, the device further includes: a parameter value determination module, a first adjustment mode determination module, and a second adjustment mode determination module.

[0094] The parameter value determination module is used to determine the parameter value of each operating parameter at the next parameter control time based on the parameter adjustment amount corresponding to at least one operating parameter at the current parameter control time and the parameter value of each operating parameter obtained in advance at the current parameter control time.

[0095] The first adjustment mode determination module is used to determine the parameter adjustment mode corresponding to the current parameter control time as direct adjustment when the parameter values ​​of each operation parameter at the next parameter control time have not reached the corresponding preset threshold.

[0096] The second adjustment mode determination module is used to determine the parameter adjustment mode corresponding to the current parameter control time as indirect adjustment when one of the operation parameters reaches a corresponding preset threshold value at the next parameter control time.

[0097] Optionally, the parameter adjustment method is direct adjustment, and the parameter adjustment module 330 includes: a valve opening adjustment unit, a reflux flow rate adjustment unit, and / or an extraction amount adjustment unit.

[0098] A valve opening adjustment unit is used to adjust the valve opening of the distillation column to be controlled based on the valve opening and the parameter adjustment amount corresponding to the current parameter control; and / or,

[0099] A reflux flow rate adjustment unit is used to adjust the reflux flow rate of the distillation column to be controlled based on the reflux flow rate and the parameter adjustment amount corresponding to the current parameter control; and / or,

[0100] The extraction amount adjustment unit is used to adjust the extraction amount of the corresponding purified object based on the parameter adjustment amount of the extraction amount of each of the purified objects.

[0101] Optionally, the parameter adjustment method is indirect adjustment, and the parameter adjustment module 330 includes: an adjustment amount display unit and an adjustment amount editing unit.

[0102] The adjustment amount display unit is used to display the parameter adjustment amount of each of the operation parameters at the current parameter control time based on the display interface of the target terminal;

[0103] An adjustment amount editing unit is configured to, in response to an edit trigger operation for at least one parameter editing item, determine a target adjustment amount corresponding to each of the operation parameters, and adjust the corresponding operation parameters based on each of the target adjustment amounts.

[0104] Optionally, the device may further include a model training module.

[0105] The operational parameter prediction model is obtained through training;

[0106] The model training module includes a training sample acquisition unit and a model training unit.

[0107] The training sample acquisition unit is used to acquire online training samples corresponding to the distillation column under control in each training round. The online training samples are determined based on the simulation environment model corresponding to the distillation column under control. The online training samples include at least one distillation process sample sequence. The at least one distillation process sample sequence includes a column temperature sample sequence and a column pressure sample sequence, a column liquid level sample sequence, a quality parameter sample sequence corresponding to at least one purification object, a liquid flow rate sample sequence corresponding to at least one target pipeline, and a reward feedback information sequence. The sample sequence values ​​included in each distillation process sample sequence correspond to the control time of each parameter.

[0108] The model training unit is used to train the operational parameter prediction model based on the simulated online training samples and reinforcement learning algorithms.

[0109] Optionally, the model training unit includes: a probability distribution determination subunit, an expected reward determination subunit, a first model parameter update subunit, a probability set determination subunit, a reward feedback information determination subunit, a second model parameter update subunit, and a model determination subunit.

[0110] The probability distribution determination subunit is used to input the tower temperature sample sequence, tower pressure sample sequence, tower liquid level sample sequence, quality parameter sample sequence corresponding to at least one purification object, and liquid flow sample sequence corresponding to at least one target pipeline in the current online training sample into the operation parameter prediction model for each of the simulation online training samples, so as to obtain the adjustment probability distribution of the distillation tower under control at the current parameter control time. The adjustment probability distribution includes at least one actual adjustment probability set, and each actual adjustment probability set includes the actual adjustment amount of each operation parameter and the corresponding probability information.

[0111] The expected reward determination subunit is used to input the actual adjustment amount of each operating parameter in the current actual adjustment amount probability set, the sample sequence of tower temperature and tower pressure corresponding to each preset tower height, the sample sequence of tower liquid level height, the sample sequence of quality parameters corresponding to at least one purification object, and the sample sequence of liquid flow rate corresponding to at least one target pipeline into the state action value model for each actual adjustment amount probability set in the adjustment amount probability distribution, so as to obtain the expected reward corresponding to the current actual adjustment amount probability set.

[0112] The first model parameter update subunit is used to update the parameters of the operational parameter prediction model based on the adjustment probability distribution, the expected rewards, and the policy gradient algorithm; and,

[0113] The probability set determination subunit is used to determine the target actual adjustment probability set based on the adjustment probability distribution.

[0114] The reward feedback information determination subunit is used to process the sample sequences of tower temperature, tower pressure, tower liquid level, quality parameters of at least one purification object, liquid flow rate of at least one target pipeline, and actual adjustment of each operating parameter in the probability set of the target actual adjustment according to the pre-set reward function, and to determine the reward feedback information of the distillation tower under control at the current parameter control time.

[0115] The second model parameter update subunit is used to update the parameters of the state-action value model based on the reward feedback information and the temporal difference algorithm, and to update the reward feedback information sequence based on the reward feedback information.

[0116] The model determination subunit is used to end training when the preset training target corresponding to the operation parameter prediction model is reached, thereby obtaining the operation parameter prediction model.

[0117] Optionally, the reward function is determined based on the observation indicator safety function, the observation indicator stability function, the control indicator anti-interference indicator function, and / or the regional value attribute indicator function.

[0118] The atmospheric distillation column control device provided in the embodiments of the present invention can execute the atmospheric distillation column control method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0119] Example 4

[0120] Figure 4 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0121] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0122] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0123] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as atmospheric distillation column control methods.

[0124] In some embodiments, the atmospheric distillation column control method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the atmospheric distillation column control method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the atmospheric distillation column control method by any other suitable means (e.g., by means of firmware).

[0125] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0126] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0127] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0128] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0129] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0130] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0131] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0132] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for controlling an atmospheric distillation column, characterized in that, include: For each parameter control time during the operation of the distillation column to be controlled, the distillation process data corresponding to the current parameter control time of the distillation column to be controlled is obtained. The distillation process data includes at least one distillation process sequence. The at least one distillation process sequence includes the column temperature sequence and column pressure sequence, column liquid level sequence corresponding to each preset column height, the quality parameter sequence corresponding to at least one purification object, and the liquid flow rate sequence corresponding to at least one target pipeline. Each sequence value included in each distillation process sequence corresponds to each parameter control time. Based on a pre-determined operating parameter prediction model, the distillation process data corresponding to the current parameter control time is processed to obtain the parameter adjustment amount of at least one operating parameter of the distillation column under control at the current parameter control time. The operating parameter prediction model is trained based on a reinforcement learning algorithm. The at least one operating parameter includes valve opening, reflux flow rate, and / or the extraction amount of each purified object. For each operating parameter, the parameter adjustment amount of the current operating parameter is the increase or decrease in the parameter value of the current operating parameter. The operating parameters are adjusted based on the parameter adjustment amount of the at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time; The method further includes: For each parameter control moment, based on the parameter adjustment amount corresponding to at least one operating parameter at the current parameter control moment and the parameter values ​​of each operating parameter obtained in advance at the current parameter control moment, the parameter values ​​of each operating parameter at the next parameter control moment are determined; If none of the operation parameters reach the corresponding preset threshold at the next parameter control time, the parameter adjustment method at the current parameter control time is determined to be direct adjustment. If one of the operating parameters reaches a preset threshold at the next parameter control time, the parameter adjustment method at the current parameter control time is determined to be indirect adjustment.

2. The method according to claim 1, characterized in that, The parameter adjustment method is direct adjustment. The adjustment of the operating parameters based on the parameter adjustment amount of the at least one operating parameter and the parameter adjustment mode corresponding to the current parameter control time includes: The valve opening of the distillation column to be controlled is adjusted based on the parameter adjustment amount corresponding to the current parameter control; and / or, The reflux flow rate of the distillation column to be controlled is adjusted based on the parameter adjustment amount corresponding to the current parameter control, and / or, The extraction amount of the corresponding purified object is adjusted based on the parameter adjustment amount of each purified object.

3. The method according to claim 1, characterized in that, The parameter adjustment method is indirect adjustment. The adjustment of the operating parameters based on the parameter adjustment amount of the at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time includes: The display interface based on the target terminal shows the parameter adjustment amount corresponding to each of the operation parameters at the current parameter control time; In response to an edit trigger operation for at least one parameter editing item, a target adjustment amount corresponding to each of the operation parameters is determined, and the corresponding operation parameter is adjusted based on each of the target adjustment amounts.

4. The method according to claim 1, characterized in that, Also includes: The operational parameter prediction model is obtained through training; The training process yields the operational parameter prediction model, including: For each training round, online training samples corresponding to the distillation column under control in the current training round are obtained. The online training samples are determined based on the simulation environment model corresponding to the distillation column under control. The online training samples include at least one distillation process sample sequence. The at least one distillation process sample sequence includes the column temperature sample sequence and column pressure sample sequence, column liquid level sample sequence, quality parameter sample sequence corresponding to at least one purification object, liquid flow sample sequence corresponding to at least one target pipeline, and reward feedback information sequence. The sample sequence values ​​included in each distillation process sample sequence correspond to the control time of each parameter. The operational parameter prediction model is trained based on the online training samples and reinforcement learning algorithms described above.

5. The method according to claim 4, characterized in that, The step of training the operational parameter prediction model based on the online training samples and reinforcement learning algorithm includes: For each of the online training samples, the sample sequences of tower temperature, tower pressure, tower liquid level, quality parameters corresponding to at least one purification object, and liquid flow rate corresponding to at least one target pipeline in the current online training samples are input into the operating parameter prediction model to obtain the adjustment probability distribution of the distillation tower under control at the current parameter control time. The adjustment probability distribution includes at least one actual adjustment probability set, and each actual adjustment probability set includes the actual adjustment amount of each operating parameter and the corresponding probability information. For each actual adjustment probability set in the adjustment probability distribution, the actual adjustment of each operating parameter in the current actual adjustment probability set, the sample sequence of tower temperature and tower pressure corresponding to each preset tower height, the sample sequence of tower liquid level height, the sample sequence of quality parameters corresponding to at least one purification object, and the sample sequence of liquid flow rate corresponding to at least one target pipeline are input into the state action value model to obtain the expected reward corresponding to the current actual adjustment probability set. Based on the aforementioned adjustment probability distribution, the expected rewards, and the policy gradient algorithm, the parameters of the operational parameter prediction model are updated; and, Based on the aforementioned adjustment probability distribution, determine the target actual adjustment probability set; Based on the pre-set reward function, the sample sequences of tower temperature, tower pressure, tower liquid level, quality parameters of at least one purification target, liquid flow rate in at least one target pipeline, and the actual adjustment of each operating parameter in the probability set of the target actual adjustment are processed to determine the reward feedback information of the distillation tower under control at the current parameter control time. Based on the reward feedback information and the temporal difference algorithm, the parameters of the state-action value model are updated, and the reward feedback information sequence is updated based on the reward feedback information. Training ends when the preset training objective corresponding to the operation parameter prediction model is achieved, thus obtaining the operation parameter prediction model.

6. The method according to claim 5, characterized in that, The reward function is determined based on the observation indicator safety function, the observation indicator stability function, the control indicator anti-interference indicator function, and / or the regional value attribute indicator function.

7. A control device for an atmospheric distillation column, characterized in that, include: The data acquisition module is used to acquire distillation process data corresponding to the current parameter control time of the distillation column under control during the operation of the distillation column under control. The distillation process data includes at least one distillation process sequence, which includes a column temperature sequence, a column pressure sequence, a column liquid level sequence, a quality parameter sequence corresponding to at least one purification object, and a liquid flow rate sequence corresponding to at least one target pipeline. Each sequence value included in each distillation process sequence corresponds to the control time of each parameter. The data processing module is used to process the distillation process data corresponding to the current parameter control time based on a pre-determined operating parameter prediction model, to obtain the parameter adjustment amount of at least one operating parameter of the distillation column under control at the current parameter control time, wherein the operating parameter prediction model is obtained by training based on a reinforcement learning algorithm; the at least one operating parameter includes valve opening, reflux flow rate and / or extraction amount of each purified object; wherein, for each operating parameter, the parameter adjustment amount of the current operating parameter is the increase or decrease of the parameter value of the current operating parameter; The parameter adjustment module is used to adjust the operating parameters based on the parameter adjustment amount of the at least one operating parameter and the parameter adjustment method corresponding to the current parameter control time; The device further includes: The parameter value determination module is used to determine the parameter value of each operating parameter at the next parameter control time based on the parameter adjustment amount corresponding to at least one operating parameter at the current parameter control time and the parameter value of each operating parameter obtained in advance at the current parameter control time. The first adjustment mode determination module is used to determine the parameter adjustment mode corresponding to the current parameter control time as direct adjustment when the parameter values ​​of each operation parameter at the next parameter control time have not reached the corresponding preset threshold. The second adjustment mode determination module is used to determine the parameter adjustment mode corresponding to the current parameter control time as indirect adjustment when one of the operation parameters reaches a corresponding preset threshold value at the next parameter control time.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the atmospheric distillation column control method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the atmospheric distillation column control method according to any one of claims 1-6.

Images